In silico antigenic site evaluation and antiviral therapy against dengue serotypes

Non Structural protein 3 (NS3) constitute protease, helicase and polymerase that are essential for dengue virus replication. The aim of the present study is to block the replication of the virus by targeting the NS3 Protein. The retrieved sequences of NS3 protein from National Centre for Biotechnology information (NCBI) shows that the antigenic sites of the protein are highly variable in all the four serotypes of dengue virus (DENV) i.e. DENV I, DENV II, DENV III and DENV IV. DENV III found to be most distantly related serotype among all the serotypes studied using UPGMA method. The 3D structure of NS3 protein was modeled using homology modeling by MODELLER 9v8. Evaluation of the constructed NS3 protein models were done by PROCHECK, WhatIf using Exome Horizon. The derived compounds of mycophenolic acid and ribavirin were docked as ligands to the constructed models of NS3 protein using Autodock 4.2 for Protein-ligand interaction study.


Introduction
Dengue disease, caused by dengue virus infection which is found to be an endemic in over 100 countries (Brinkworth et al., 1999).It was found that 100 million cases of dengue fever occur annually.Of which, 500,000 cases require hospitalization, and 25,000 are fatal (Gubler et al., 1998;Ligon et al., 2005;Gratz et al., 1999;Halstead, 2007), due to limited healthcare facilities in developing and underdeveloped countries.DENV (Dengue virus) is an arthropod-borne flavivirus that comprises four distinct serotypes (DENV I, DENV II, DENV III and DENV IV) that constitute an antigenic complex of the genus flavivirus, family Flaviviridae (Mason et al., 1990;Henchal et al., 1982;Gentry et al., 1982;Monath et al., 1986;Russell et al., 1967).Every step in the life cycle of the dengue virus is a potential target for inhibiting viral replication (Qi et al., 2008).NS3 protein constitutes protease, helicase and polymerase that are essential for dengue virus replication (Bera et al., 2008).NS3 is responsible for proteolysis of den [gue viral RNA polyprotein as well as carrying out various enzymatic reactions that are mandatory for replication of dengue virus (Luo et al., 2008).Currently there is no antiviral therapy available for Dengue (Muhamad et al., 2010).We assessed the ability of myco -phenolic acid (MPA) and ribavirin (RBV), drug currently used as an immunosuppressive agent, to inhibit dengue virus antigen expression, RNA replication, and virus production (Diamond et al., 2012;Allison et al., 1993;Koff et al., 1983;Conner et al., 1984).The aim of this study was to examine the mutation in antigenic site of dengue virus and the antiviral action of mycophenolic acid (MPA) and ribavirin (RBV) on NS3 proteins DENV I-IV determining the best drug that can be most active against the virus from the binding energy and the pocket which fits the drug.

Sequence retrieval
The nonstructural protein 3 sequence of DENV I, DENV II, DENV III and DENV IV were obtained from the National Centre for Studies in Biotechnology (http:// www.ncbi.nlm.nih.gov).The numbers of sequences were found to be 25, 17, 103 and 25 for DENV I, DENV II, DENV III and DENV IV respectively.

Antigenic site finding
The Exome Horizon antigenic site finder tool was used to find the antigenic sites of protein.The accession No., sequence length, hits, positions, antigenic sites, antigenic site length and the highest score were given in the Table I-IV.

Phylogenetics analysis
The phylogenetic trees were also built for analyzing the highly variable sequences and to find the conserved domains of the non structural protein sequences by using UPGMA method using Exome Horizon.

Sequence alignment and homology modeling
The amino acid sequences of NS3 of DENV were taken from the National Center for Biotechnology Information (NCBI) Database (Wheeler et al., 2008).A modified Needleman and Wunsch (1970) method implemented in Exome Horizon and the BLOSUM62 substitution matrix (Henikoff et al., 1992) was applied for aligning the sequences.
The homology models of DENV NS3 were constructed applying the MODELER 9v8 program using same template structures for each domain.All the models (M1, M2, M3, and M4 for DENV I, DENV II, DENV III and DENV IV respectively) were generated using the Crystal Structure of the NS3 Protease-Helicase from Dengue virus as template structures (2VBC) (Bhattacharya et al., 2008).

Ligand preparation
The Molecules (In this study referred as ligands) were prepared using moldraw tool of Exome Horizon.The different parameters of the ligands were studied given in the Table V and VI.

Molecular docking study
The docking of ligands to the catalytic triad of NS3-NS2B protease was performed using AutoDock 4.0 software (http://autodock.scripps.edu/).AutoDock is reported to be a very common docking program (Sousa et al., 2006) and is reliable (Hetényi et al., 2002).Using the software, polar hydrogen atoms were added to the enzyme and its nonpolar hydrogen atoms were merged, whereas for the ligand, nonpolar hydrogen atoms were merged and Gasteiger charges were added.All rotatable bonds of ligands were set to be rotatable.All calculation for protein-fixed ligand-flexible docking was done using the Lamarckian Genetic Algorithm (LGA) method.A population size of 150 and 10 millions energy evaluations were used for 100 search runs.The grid box with a dimension of 60 × 60 × 60 points and 0.375 Å grid spacing was used around the catalytic triad to cover the entire enzyme binding site and allow accommodated ligands to move freely.After the docking searches were completed, clustering histogram analysis was performed based on an RMSD (root mean square deviation) of not more than 1.5 Å.The best conformation was chosen from the most populated cluster with the lowest docked energy.The interactions of complex enzyme-ligand conformations, including hydrogen bond and other interactions, were analyzed using Viewerlite (www.accelrys.com)and UCSF Chimera software (www.cgl.ucsf.edu/chimera/)(Figure 1-4).

Results
The antigenic sites analyzed from all the serotypes were found to be highly variable.In DENV-I the sites were found to be IVGLYGNGVVTTSGTYVSPIAQAK, MRLLSPVRVP, EVQVIAVE, MAVGIVSILLSSLLKND VPLAGPLIAGGMLIACYVISG, FTVVVGDVVGILAQ and HGTVLVQVKY.It is observed that the amino acids present at the site producing highest scores were found to be A, P, V, C, G and Q for DENV I.For DENV-II, the antigenic sites were found to be EVQVLALE, MRLLSPVRVPNYNLI, QLGQVMLLVLCVTQVLM, RYLPAIVREA, TSLSVSLVLVGIVTLYLGVMVQAD, and QLGQVMLLVLCVTQVLM.The amino acids producing highest scores for DENV-II were found to be V, P and L. However DENV-III showed much conserved antigenic sequences and was found to be KYLPAIVREA, DRVIDPRRCLKPVILT, and ADRVI DPRRCLKPVILTD.Proline was found to be the most conserved antigenic site for DENV-III.The DENV-IV found to be having the most similar antigenic sites as of DENV-III.The sites were found to be IVDLMCHAT, GG, and GRVIDPRRCLKPVILT.The high scoring regions were found to be L, P, E and I. Hence, the antigenic site P is found to be the most conserved region among all the four serotypes, suggesting as the potent inhibition point of the ligands.The geometry and stereochemistry of the models were evaluated by using the program PROCHECK (Sousa et al., 2006).The stereochemical quality of the generated homology models as well as the crystal structure was evaluated using Ramachandran plots.Results revealed that 79.7 and 2.4% of the residues of the models are located in the most favored regions and the additional allowed regions for Model1.Similarly the residues present in favored region and disallowed regions were found to be 88.6 and 0.2, 87.1 and 11.2, 89.3 and 0.4% for Model2, Model3 and Model4 respectively.The obtained results indicate that all models possess sufficient stereo-     chemical quality.The Phylogenetic analysis showed the high intraspecies variation in NS3 protein of all the serotypes (Figure 5).The distances within group was found to be 0.05, 1.27, 0.70 and 0.20 for DENV-I, DENV-II, DENV-III and DENV-IV respec-tively.The mean distances within groups showed that DENV-I has high distance related from DENV-II, DENV-III and DENV-IV.The homology models of the NS3 protein of four serotypes also showed very good variation in Ramachandran plot.The docking study of the derived compound of mycophenolic acid (4 nos.) and ribavirin (5 nos.) was done with AutoDock4.2.The binding energy obtained in the range of -0.97 to +190.03.The ligand 3 of mycophenolic acid was found to be the best drug in Model1 having the minimum binding energy -9.2 kcal/mol.The ligand 4 of the ribavirin revealed as the best inhibitor for model2 and showed the minimum binding energy of -16.5 kcal/mol and was found to be the best ligand among all the ligands studied.The number 4 ligand of the mycophenolic acid found to be second best inhibitor for model3 among all the selected ligands after the ligand4 of mycophenolic acid with binding energy -15.9 kcal/mol.The ligand1 of ribavirin was found to be the best inhibitor for model4 having minimum binding energy -10.7 kcal/mol (Figure 6).

Discussion
Dengue fever epidemics has increased numerously over the last few decades (Ligon et al., 2005) therefore Developing antiviral drug and vaccine is becoming very important due to the global threat of viral disease pandemics (Noble et al., 2010;Wang et al., 2009).The functional similarity between the NS2B/NS3 proteases from the four genetically and antigenically distinct serotypes was identified by the differences in their substrate specificity using tetrapeptide and oc-  tapeptide libraries in a positional scanning format, each containing 130,321 substrates (Guzman et al., 2010).Development of new genomic and proteomic studies coupled with computational sciences could provide the discovery of various target proteins and potential inhibitor to be developed as drugs (Li et al., 2005;Tambunan et al., 2011).The NS3 enzyme of dengue is responsible for replication of the virus.The replication complex include the NS3 nucleotide the NS3 protease and with its NS2B cofactor, the NS3 nucleotide triphosphatase.This protein serves as the potential inhibitory targets for antiviral agents since they are required for virus replication.The multifunctional Cterminal domain of NS3 encodes NTPase, helicase and RTPase activities.NTP hydrolysis is thought to provide the chemical energy required for helicase activity.There is currently no antiviral therapy available against dengue virus.In this study we generated some antiviral ligands of mycophenolic acid and ribavirin.We predicted the binding activity of the ligands against NS3 protein of Dengue virus.From the interaction study of the ligands it was observed that the ligands 3, 8, 4 and 5 are the best inhibitors for DEN-1, DEN-2, DEN-3 and DEN-4 respectively.These ligands showed the binding energy of -9.2, -16.5, -15.9 and -10.7 Kcal/mol respectively for the corresponding serotypes.To study the interaction of ligands at the inhibition point, it is necessary to know the antigenic sites of the disease causing protein.It is observed that, there is great variation in the antigenic site of the proteins.However the most conserved antigenic site found to be the aminoacid, Proline almost in every serotype.The ligands were docked into each antigenic site of the protein and the binding energy was reported.Phylogenetic analysis also suggested the typical variations in all the NS3 sequences of NCBI.Finding antigenic site of a mutated protein is the primary aim of any computer aided drug design.The current research discovered several variations in the original sequences as well as antigenic site in the NS3 protein of the 4 serotypes.The antigenic sites were targeted to block the replication activity of the virus.The antigenic sites were targeted by the derivative compounds of mycophenolic acid and ribavirin to block the replication process of the virus.
Hence the current study may be useful in designing the drugs that may be synthesized in wet lab and can be used as antiviral drugs against dengue serotypes.The docking of two groups of inhibitors from mycophenolic acid and ribavirin again NS3 were carried out.In this work, the complexation energy of the docking was used as the descriptors for selecting new candidates for competitive dengue inhibitors.The antigenic sites were highly variable.The phylogenetics studies were also evaluated for finding the interspecies variation in NS3.Homologies of protein were constructed for all the serotypes.AutoDock4.2helped in carrying out the interaction of the drugs with the protein models.
Complexation energies for all the new ligand-enzyme complexes were evaluated.Detailed structural information is becoming increasingly available for the dengue NS3 proteins.Since these proteins are requested for virus infectivity and replication.Structure based computational approaches offer an attractive strategy for the discovery and optimization of dengue antiviral drugs.Moreover, these computational approaches promise and improve the effectiveness of current structure based calculations.

Conclusion
Based on the complexation energies calculated, the 3 rd compound of the mycophenolic was found to be the best inhibitor for DENV I. Similarly drug4 of ribavirin, drug4 of mycophenolic acid and drug1 of ribavirin were found to be best drugs against DENV II, DENV III and DENV IV respectively the lowest and closest energies to the reference compounds.

Figure 1 :Figure 4 :
Figure 1: Docking Interaction of Models with ligand1 and 2. The interaction energy was calculated and the hydrogen bonds were observed

Figure 6 :
Figure 6: Phylogenetic analysis of DENV I